Decentralized blockchain governance for explainable causal AI in macroeconomic forecasting

Abstract

The increasing occurrence of structural breaks, financial crises, and economic uncertainty has significantly weakened the predictive performance of traditional macroeconomic forecasting models. In response to these limitations, this study proposes a blockchain-enabled causal machine learning framework designed to improve forecasting robustness, transparency, and interpretability under unstable economic conditions. The proposed framework integrates blockchain-based data governance mechanisms, structural break detection techniques, machine learning algorithms, causal inference methodologies, and Explainable Artificial Intelligence (XAI) tools within a unified forecasting architecture. More specifically, the study combines Bai–Perron structural break analysis, Markov-Switching models, Double Machine Learning Causal Forest estimation, and SHAP-based explainability techniques to capture nonlinear and regime-dependent macroeconomic dynamics. The empirical analysis relies on a large macroeconomic dataset covering major crisis episodes, including the 2008 global financial crisis, the COVID-19 pandemic, and the 2022 inflationary shock period. The forecasting performance of the proposed framework is compared with conventional econometric models and standard machine learning algorithms such as VAR, TVP-VAR, Random Forest, XGBoost, and LSTM networks. The results demonstrate that the Double Machine Learning framework significantly outperforms benchmark models across different forecasting horizons and uncertainty regimes. The findings further reveal that uncertainty indicators, oil prices, financial volatility, and monetary policy variables exert strong regime-dependent effects on inflation dynamics. In addition, the proposed framework incorporates a blockchain-based governance layer designed to enhance data provenance, auditability, reproducibility, and decentralized governance throughout the forecasting pipeline. While the empirical evaluation focuses on forecasting performance, the blockchain layer provides the architectural foundation for secure and transparent model lifecycle management. Overall, the study contributes to the emerging literature at the intersection of blockchain technologies, causal artificial intelligence, and macroeconomic forecasting by developing a decentralized and explainable forecasting framework capable of supporting adaptive policy analysis under uncertain economic environments.